Keith Devlin, Stanford University
The US ranks much worse than most of our economic competitors in the mathematics performance of high school students.
We now have the knowledge to turn that around. We could raise the level of mathematics performance across the board, within a single school generation, so that we are number one in the world. All it would take is a one-time, national investment of $100 million over a five-year period. That’s what it would cost to build and put in place a system that could achieve that change, with the existing school system and the existing teachers. Once built, that system would be self-sustaining.
That sounds like a lot of money for an upfront investment. But thought of as a national initiative, it’s peanuts. The payoff for the nation’s health and future prosperity is far greater than the long term benefits we got from the far greater investment in NASA’s Apollo Program to put a man on the Moon.
I don’t think it’s going to happen in this way, but not because people don’t think it’s a good idea. Rather, it would probably require a combination of nonprofit and for-profit funding that our system does not allow.
The same goal can, and surely will, be attained. But it will take a
lot longer.
I’ll tell you, briefly, what the approach is, how I am so sure it will
work, and where I got that cost figure. Everything I say is based on
work that has already been done.
First, let me tell you who I am.
I’m a mathematician at Stanford who directs a multidisciplinary think tank called the H-STAR institute, that looks at issues involving human sciences and new technologies, with a view to improving technology design and use, including applications of technology in education at all levels. (I’m also the Math Guy on National Public Radio.)
What I want to tell you about is connected with the H-STAR institute, but is based on some work I’ve just completed as an individual, working with a large software company in Silicon Valley.
We have spent the past four years looking to see if we can use the range of today’s technologies to improve the dismal math performance level of the nation’s high school students.
The slide in math performance among US children occurs during the age range 8 to 13. Essentially the middle-school years. That was the target group for our study.
Many attempts have been made to improve US middle-school mathematics education, but all have failed to achieve the desired results. I think the reason is clear. They have all focused on improving basic math skills.
In contrast, I (and a great many of my colleagues) believe the emphasis should be elsewhere. Mathematics is a way of thinking about problems and issues in the world. Get the thinking right and the skills come largely for free.
There are two reasons why the focus has been on skills. First, many people, even those in positions of power and influence don’t understand what mathematics is and how it works. All they see are the skills, and they think, wrongly, that is what mathematics is about. (Given that for most people, their last close encounter with mathematics was a skills-based school math class, it is not hard to see how this misconception arises.)
The other reason is more substantial. For over two thousand years, the only way to provide mathematics education to the masses was through the written word. Textbooks. But in order to learn mathematical thinking from a textbook, you have to approach it via the skills. That means you have to master the skills first.
But as I already remarked, mathematics is not about acquiring basic skills or learning formulas. It’s a way of thinking. It’s not about things you know, it’s something you do. And the printed word is a terribly inefficient way to learn how to do something.
The best way for an individual to learn how to do something is, as the Nike slogan says, “Just do it!”
Until now, learning by doing was not a viable approach to mathematics education. It was possible one-on-one, by an apprenticeship system, but not on a broad scale. Now it can be done.
We now have the know-how to raise the mathematical performance of our nation’s schoolchildren in the 8 to 13 age-range to the top of the world rankings in a single school generation.
The method is simulation. That’s the way we train pilots to fly aircraft, the way we train astronauts to fly the shuttle and to work in the Space Station, the way we train surgeons, and the way the US Army trains soldiers before they go anywhere near the battlefield.
And that’s the way we should train young people to think mathematically.
The technology to do that has been provided to us by the leisure and entertainment industries. Basically, it’s videogame technology and Web 2.0 infrastructure.
No one has yet tried to do this on the scale that is required. Yes, there are a lot of so-called math ed videogames out there. Lots of them are very superficial, some are more thoughtfully designed. But they all focus primarily on skills. They use the compelling nature of videogames as a wrapper for conventional curriculum, to try to get kids to learn and practice the basic skills. But as I’ve noted, mastery of skills does not lead to mathematical thinking.
For over two thousand years, mastery of mathematical skills had to come before developing the higher level thinking because we did not have simulators. All we had was books. Now we know how to build simulators.
Based on the work I and my colleagues have done over the last four years, we have a pretty good sense of what it would take to build such a simulator. That’s where I get my figure of $100 million over five years. Building the simulator in the first place would cost around $50 million. (That was the cost of building World of Warcraft.) The remaining amount is what it would cost to build the infrastructure to support and maintain the system for use across the nation. Once in place, it could be self-sustaining through user subscriptions.
Maria,
Has he posted the statement below somewhere public? I think it's a mistake to not address what happens to kids when they're younger.
I'll try to think of some questions...
Warmly,
Sue
The method is simulation. That’s the way we train pilots to fly aircraft, the way we train astronauts to fly the shuttle and to work in the Space Station, the way we train surgeons, and the way the US Army trains soldiers before they go anywhere near the battlefield. And that’s the way we should train young people to think mathematically.
Mathematics is a way of thinking about problems and issues in the world. Get the thinking right and the skills come largely for free.
Hello,
Several Math 2.0 members want to invite Keith to our weekly series and talk to him live about a math program his think tank envisions. I think we could have a valuable conversation, especially about OERs, grassroots movements, and blended financing. I would like everybody who reads this to ask a question they would ask Keith, as an experimental pre-meeting activity. One value this network has is in constructive, on-topic, deep questions we ask.
So, QUESTIONS PLEASE.
On May 21, 2010, at 3:12 PM, Maria Droujkova wrote:I don't think he's looking for one centralized platform, but rather a place where folks want to go to learn stuff in an interesting way and doing math is a big part of it - like Google is to search... but then they are trying to become one centralized platform... a little like what Apple is turning into. The math 2.0 simulator does not have to rule the world, just become an important object in it.
> And a question: Why do we need one centralized Platform? As opposed to using web as a platform, with all the wealth of stuff there, including giants like Warcraft and Eve, and all the way to little Scratch applets a kid can write in an afternoon...
-Ihor
Some of the opposition forces are already starting to gather in the wings pilling up their ammo...
I hope you don't think I am in opposition (need a winky/smiley face after a statement like that :)) I just think that anyone who wants the public or government to fund a $100 million project should anticipate a few challenging questions. The questions I asked were things that I had thought about in relation to my own projects. I spent over a year developing math simulations in Second Life for middle school students back in 2007. A lot of educators were really excited about the possibilities but three years have passed and nothing much came of it. I know Marc Prensky was developing an algebra game whose tagline was "play the game, pass the course". That was at least 5 years ago and the project has stalled indefinitely. There are a lot of great ideas that just never come to fruition.
I fully support the idea of a math simulator. In fact, I would even love to work on the project in some capacity. I want Keith Devlin to be a guest so we can all explore this idea more fully. I hope he is open to such a conversation. I'm not optimistic that he will be but I'm anxious to be proved wrong.
Rob Tucker on O'Reilly Radar makes an important point about ownership, and there's a lively discussion:
http://radar.oreilly.com/2010/04/ed-20-the-importance-of-owners.html
Bryce, in comments, calls Keith's idea "World of Mathcraft" and is concerned about aiming for extrinsic tools like progress tracking and achievements (the bane of World of Warcraft of late, alas).
Hiya,
Maybe this is not the place for this. Just ignore. But I have been thinking so hard about this and didn’t know where to post.
Something that always confused me in school was the connection between Function Rule, Table of Values, Graph - primarily because we use a “generic” format and yet each problem has its own restrictions. I particularly hated the term “Table of Values”, because usually it was just “Table of SOME Values”.
==
Problem 1: My boss gives me 5 shirts to sell. I make a profit of $30 per shirt. http://geogebrawiki.pbworks.com/Modeling_ShirtProblem
Problem 2: I drive 5 hours at 30mph. http://geogebrawiki.pbworks.com/Modeling_DriveProblem - I love the dynamic table and it is easy to make!
==
I am thinking that GeoGebra gives us the ability to write “complete descriptions” since we have
(a) Algebra View so we can define rules and functions, limit them to intervals, ….
(b) Graphics View and we can label the axis, the units, etc.
(c) Spreadsheet View and we can make dynamic tables of values, convert automatically to lists and get idea of function.
==
Yesterday someone linked me to: http://www.dec.utexas.edu/askme/unit1.html#activity10 which is called Four-Corner Model.
QUESTIONS:
1. Would the answer to the Four Corner Model for these 2 problems be different?
2. Do kids get understand that tables are sometimes complete and sometimes not and that functions sometimes give bad information (like selling 1/2 a shirt) ?
3. What do you think of “Descriptions” with technology like GeoGebra?
I would be grateful for any insight/ experience/ feedback etc.
Linda
--------------------------------
Goal: To create GeoGebra Challenge for modeling (see e.g. http://www.youtube.com/watch?v=9th6fvaO3EM )
2. Do kids get understand that tables are sometimes complete and sometimes not and that functions sometimes give bad information (like selling 1/2 a shirt) ?
Students who use the Rule of Four are more likely to make that connection than those who do not. I really like how you show this in your examples.
3. What do you think of “Descriptions” with technology like GeoGebra?
I think this is brilliant and should be explored further. I would encourage you to apply this to other types of problems and, perhaps, give a presentation.
Colleen
________________________________
To unsubscribe from this group, send email to mathfuture+...@googlegroups.com <mailto:mathfuture%2Bunsu...@googlegroups.com> .
Hiya Colleen,
Thank you so much for your detailed response – that was exactly what I was hoping for so that I could see if I was looking at this problem in a useful way.
I will think about this more. In 8th grade I had this fantastic teacher who made us write with words: Let x=# shirts sold. Then x can be 0,1,…,5. on one line and then the next line was: Let y=profit in $. Then y=30x. Everything had to have words and units. It was SUCH a bore – but I swear it was the single most important tool I learned.
(P.S. I did not even think about the fact that problems are often worked backwards and forwards as with your response to Q1.)
Thanks again,
Linda
From: mathf...@googlegroups.com [mailto:mathf...@googlegroups.com] On Behalf Of Colleen King
Sent: Tuesday, May 25, 2010 8:37 PM
To: mathf...@googlegroups.com
Hi Linda,This is most certainly the right place for your question.I hope to see more questions like this.Around my area, we refer to this as the Rule of Four or simply "multiple representations". Some teachers ask students to fold a piece of paper into fourths and describe the problem differently in each quadrant. I think Geogebra offers a better way for students to model math problems.